Expand description
Gaussian Process regression (PRML ch 6.4) — a nonparametric Bayesian regressor
that returns a full predictive distribution (mean, variance) at every input,
with the squared-exponential (RBF) kernel. The training solve reuses
linear_algebra::cholesky (no new solver). Kernel-class DenseLinear (the
n×n kernel solve) + AllPairs (the kernel evaluations).
Given training (X, y) and noise variance σ²ₙ:
mean(x*) = k*ᵀ (K + σ²ₙI)⁻¹ y,
var(x*) = k(x*,x*) − k*ᵀ (K + σ²ₙI)⁻¹ k*,
the calibrated uncertainty that collapses near training points and widens away
from them.
Structs§
- GpRegressor
- A fitted GP regressor (squared-exponential kernel).